Using longitudinal data, the proposed study will investigate the direct and indirect effects of continuing and evolving family, social, and economic conditions on physical and psychological well-being in a cohort of African American women originally recruited in 1966 when their children were first graders. The data come from a community epidemiological study of children and their families initiated in the 1960s in Woodlawn, a poor African American community on the south side of Chicago. The children and their mothers were followed over time, and the proposed study focuses on the mothers (n=1136). The women were interviewed at three time points over 25 years: 1967, when their children were in first grade (median age of 30); 1975, when their children were adolescents; and 1997, when their children were adults. In 1997, 1008 (89%) of the original cohort of women were located, and their ages ranged from 49 to 79 with a median age of 60. Most of the women were past the traditional childrearing stage of life and were approaching or had made the transition to retirement. They reported a wide array of physical and psychological health conditions as well as a number of functionallimitations and other disabilities. [unreadable] Our conceptual framework is based on a life course developmental perspective that focuses on roles within major social fields across the life course. The focus of the proposed study is to delineate the patterns of roles across the major social fields of family, work, and community, and to identify the pathways from these roles in early (median age 31, 1967) and mid adulthood (median age 39, 1975), to physical and psychological well being in later life (median age 61, 1997). The proposed research will require several different analytic approaches. We will begin with exploratory analyses to describe univariate distributions then we will conduct bivariate analyses to assess relationships among the variables. Our multivariate techniques will include regression analysis, survival analysis, transition analysis (latent class and latent transitional modeling), structural equation modeling, latent variable mixture modeling, and generalized estimating equations (GEE). We will study how family, work, and community in early and mid adulthood influence health of older women. Little is known about family and social influences on later health. This study will provide knowledge about African American women, an understudied group.